Patients with a history of idiopathic gastroduodenal ulcer bleeding face an increased risk of recurrent ulcer gastrointestinal bleeding. Our ongoing clinical trial demonstrates a possible reduced risk of recurrent idiopathic gastroduodenal ulcer bleeding with proton pump inhibitor (PPI), yet there is a significant risk of recurrent ulcer bleeding as PPI may increase the risk of small bowel bleeding. Our preliminary data provide strong plausibility that a combination therapy of misoprostol (MISO) with a PPI reduces the recurrent ulcer bleeding as well as clinical gastrointestinal bleeding. The investigators are going to provide the definitive answer to this important clinical question through a randomised trial.

HYPOTHESIS The hypothesis that a combination therapy of misoprostol and lansoprazole is superior to lansoprazole alone for the prevention of recurrent ulcer bleeding in patients with a history of idiopathic ulcer bleeding.

STUDY DESIGN OVERVIEW It is a two-year, double blinded, randomized trial of a combination therapy with misoprostol with PPI (lansoprazole) versus misoprostol placebo plus PPI (lansoprazole) in patients with a history of idiopathic ulcer bleeding.

Randomization All eligible patients will be randomly assigned (in a 1:1 ratio) to receive 24 months of either misoprostol 800 micrograms daily (i.e. misoprostol 200 micrograms four times daily) combined with lansoprazole 30 mg once daily, or misoprostol placebo four times daily plus lansoprazole 30 mg once daily. A computer-generated randomisation schedule is used to assign patients to the treatment sequences. Concealment of allocation will be ascertained by an independent research staff member. The Clinical Research Pharmacy located in Prince of Wales Hospital will dispense consecutively numbered, identical packs that contain sealed bottles of the study medications.

Follow-up assessment After the randomisation visit, patients will return at month 2, month 6, and then every four months thereafter until 24 months (follow up ± 14 days from scheduled clinic visit is allowed). At each visit, we will assess patients' complete blood picture, renal and liver function tests, and serum salicylate level, compliance to study medications, and the use of other medications including over-the-counter drugs, and safety of the treatment. Drug adherence is assessed by counting the study drugs. Patients are permitted to take antacids to relieve dyspepsia. Drugs prohibited during the study include anticoagulant agents, NSAIDs, cyclooxygenase-2 inhibitors, over-the-counter analgesics (including herbal products), misoprostol, sucralfate, antiplatelet drugs, bisphosphonates, and PPIs/H2RAs apart from the study drugs.

:

Inclusion Criteria:

1. Idiopathic gastroduodenal ulcer bleeding is defined as described in our previous studies [12-15]:

   Gastroduodenal ulcer bleeding is diagnosed based on:

   i. History of symptoms of upper gastrointestinal bleeding with endoscopically proven gastroduodenal ulcers;

   Idiopathic ulcer is diagnosed based on:

   i. No experience of ulcerogenic agents (e.g. aspirin, NSAIDs), or drugs of an unknown nature including traditional Chinese medicine during the 4 weeks before hospitalization of gastrointestinal bleeding episode; ii. Negative biopsy urease test and absence of H. pylori on histology in the absence of acid suppressive agents; and iii. No other cause of ulceration identified (e.g. hypergastrinaemia, Crohn's disease, cytomegalovirus and herpes infection).

2. Resume hemoglobin level which is same as or higher than the level prior to last ulcer bleeding episode or stable hemoglobin level (drop <2g/dL) within one year prior randomization iii.3. No reported gastroduodenal ulcer or ulcer bleeding from last upper endoscopy 2.4. Aged 18 years old or above. 3.5. Written informed consent obtained. Fingerprint of subject with a witness involved in the consent procedure will be accepted for illiterate subjects.

Exclusion Criteria:

Patients will be excluded from the study if they have any of the followings:

1. Concomitant anticoagulant
2. Concomitant use of NSAIDs, aspirin or COX2 inhibitors
3. Previous gastric surgery
4. Requirement of maintenance PPI (e.g. reflux esophagitis)
5. Advanced comorbid conditions (defined as American Society of Anesthesiologists grade 4 or above) or active malignancy
6. Subjects who are or will be pregnant or lactating
7. Subjects who have known hypersensitivity or allergies to any component of misoprostol and lansoprazole.
8. Subject who has current or historical evidence of hypergastrinaemia syndrome or other hypersecretory condition.